When using python2.7.10
When I tried using such a script, The result is that the processing speed of map is faster if it is simple. (Memory usage is delicate)
import sys
@profile
def which_is_good():
    print sys.version
    l = range(100000)
    mapped = map(lambda d: chr(65 + d%127), l)
    m_stringed = str(mapped[:])
    del mapped
    del m_stringed
    comped = [chr(65 + d%127) for d in l]
    c_stringed = str(comped[:])
    del comped
    del c_stringed
    mapped, comped = (), ()
which_is_good()
kernprof -vl run.py
2.7.10 (default, Jul 13 2015, 12:05:58) 
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)]
Wrote profile results to run.py.lprof
Timer unit: 1e-06 s
Total time: 0.155445 s
File: run.py
Function: which_is_good at line 3
Line #      Hits         Time  Per Hit   % Time  Line Contents
==============================================================
     3                                           @profile
     4                                           def which_is_good():
     5         1           33     33.0      0.0      print sys.version
     6                                           
     7         1         2086   2086.0      1.3      l = range(100000)
     8                                           
     9         1        49070  49070.0     31.6      mapped = map(lambda d: chr(65 + d%127), l)
    10         1        21482  21482.0     13.8      m_stringed = str(mapped[:])
    11         1          150    150.0      0.1      del mapped
    12         1           23     23.0      0.0      del m_stringed
    13                                           
    14    100001        68510      0.7     44.1      comped = [chr(65 + d%127) for d in l]
    15         1        13911  13911.0      8.9      c_stringed = str(comped[:])
    16         1          152    152.0      0.1      del comped
    17         1           27     27.0      0.0      del c_stringed
    18                                           
    19         1            1      1.0      0.0      mapped, comped = (), ()
python -m memory_profiler run.py
2.7.10 (default, Jul 13 2015, 12:05:58) 
[GCC 4.2.1 Compatible Apple LLVM 6.1.0 (clang-602.0.53)]
Filename: run.py
Line #    Mem usage    Increment   Line Contents
================================================
     3   19.551 MiB    0.000 MiB   @profile
     4                             def which_is_good():
     5   19.555 MiB    0.004 MiB       print sys.version
     6                             
     7   22.660 MiB    3.105 MiB       l = range(100000)
     8                             
     9   23.426 MiB    0.766 MiB       mapped = map(lambda d: chr(65 + d%127), l)
    10   28.496 MiB    5.070 MiB       m_stringed = str(mapped[:])
    11   28.246 MiB   -0.250 MiB       del mapped
    12   27.996 MiB   -0.250 MiB       del m_stringed
    13                             
    14   28.730 MiB    0.734 MiB       comped = [chr(65 + d%127) for d in l]
    15   28.840 MiB    0.109 MiB       c_stringed = str(comped[:])
    16   28.590 MiB   -0.250 MiB       del comped
    17   28.340 MiB   -0.250 MiB       del c_stringed
    18                             
    19   28.340 MiB    0.000 MiB       mapped, comped = (), ()